The influence of reduced audible bandwidth on asynchronous double-vowel identification

Abstract

Purpose: In this study, the authors sought to determine whether reduced audible bandwidth associated with hearing loss contributes to difficulty benefiting from an onset asynchrony between sounds.

Method: Synthetic double-vowel identification was measured for normal-hearing listeners and listeners with hearing loss. One vowel (Target 2) was 250 ms in duration, and one (Target 1) varied in duration. The vowels had the same offset, and an onset asynchrony between the vowels ranged between 0 and 200 ms. Listeners identified both vowels in their perceived order. The scoring metrics used were as follows: Target 1 correctly identified in the correct position, Target 2 correctly identified in the correct position, ordered double-vowel identification, and unordered double-vowel identification. The same experiment was conducted with vowels low-pass filtered at 900 Hz simulating reduced audible bandwidth.

Results: For all scoring metrics, increases in onset asynchrony led to better vowel identification. Listeners with hearing loss benefited less from onset asynchrony than normal-hearing listeners only for Target 2 identification. Filtering the vowels reduced onset asynchrony benefit for all scoring categories and for both groups.

Conclusion: Results implicate reduced audible bandwidth in difficulties of listeners when using onset asynchrony for sound segregation. Therefore, listeners with a reduced audible bandwidth may have communication difficulties in natural environments.